I have a couple of questions and comments about the draft:
1) Was there a technical reason for the new packet length format.
Limiting packets larger than 8383 bytes to powers of two seems a bit
restrictive. Also, if that scheme is left in place, you should describe how
padding is done and how to find the end of useful data.
2) It is not clear whether this document is supersceding or
supplementing RFC1991. The introduction suggests it is supplementing. In that
case, I would not repeat anything from RFC1991 so as not to 'blimp-out' this
document. One place you do this is in describing the CTB and packet length
formats. If this is a supplement, then simply refer the reader to RFC1991 since
this format is not generated by newer PGPs. Note that this does not apply to
elements that are still valid for the new PGP formats. For instance, you must
still define all of the packet types that are still in use with the new PGP.
3) I also agree with earlier comments on the list that all currently
supported packet types should be fully documented, not just 'some of the more
interesting packet types'.
4) It would be really helpful to use ascii-character box diagrams to
illustrate the CTB, packet length, and the assorted packet layouts (ala older
IP/TCP RFCs). They look kind of old and crufty but can clear up a great deal of
confusion when done correctly.
--
Tony Mione, RUCS/NS, Rutgers University, Hill 055, Piscataway,NJ - 732-445-0650
mione(_at_)nbcs-ns(_dot_)rutgers(_dot_)edu W3:
http://www-ns.rutgers.edu/~mione/
PGP Fingerprint : E2 25 2C CD 28 73 3C 5B 0B 91 8A 4E 22 BA FA 9F
Editorial Advisor for Digital Systems Report ***** Important: John 17:3 *****